• by September 29, 2011

Filtering is the dominant media problem of the age. Set yours too coarsely, and you drown in distractions; set it too finely, and you miss the Big Idea.

Good filters combine and balance different approaches. Consider the New York Times' set: professional (Tom Friedman), social (most emailed), tribal (what your Facebook friends recommend). To these, Google News adds the algorithmic ("Personalize" topics and sources).

William Gibson fans love quoting "The future's already here, it's just not evenly distributed." We're in the middle of an interesting transition, from filters as stubby features growing out of content properties, as with the Times, toward platforms purpose-built as filters. Witness the newly-launched harvard.edu, reconceived as an aggregator/syndicator of content from the many large and small corners of the university. And for you Minority Report pre-cog police devotees, check out darwineco.com.

To me, today's best filters pre-aggregate a well-sifted (or as the digerati say these days, "curated") content corpus, and then give you a little control over it. In the spirit of "show, don't tell", here's a little present: I used Yahoo! Pipes (a simply brilliant service I hope doesn't wither on the sad vines there) to create a keyword-filtered subscription service out of MediaPost's main feeds (Online, Media, and Marketing), so you can pluck the things you care about most out of all this wisdom. Enjoy: bit.ly/p5I4j6 (Try "Ad Networks", for example.)

What lies beyond? We're beginning to add structure and meaning to the content we publish -- that "Semantic Web" thing you've been hearing about. With specifications like SPARQL (here's a good tutorial: http://www.cambridgesemantics.com/2008/09/sparql-by-example/) we'll be able to define, store, and run "queries" across well-tagged media. So the future will enable something like, "show me stuff about 'filtering' that more than three of the people Cesar follows on Twitter re-tweet."

So, if you're a media executive or entrepreneur, or just a nerd looking for a good meaty problem, for my money this filtering thing's got legs, and it's still early days.

Cesar A. Brea, Partner, Force Five Partners, LLC